Information, work, and heat in small systems

大约 150 年前，詹姆斯·克拉克·麦克斯韦 (James Clerk Maxwell) 首次提出了当“灵巧生物”（恶魔）利用信息从热量中提取有用能量（做功）时会发生什么。热力学第二定律应该禁止这样做，但是怎么办呢？这个问题让物理学家感到恼火，并引发了一个多世纪的争论。 从 2008 年开始，我的实验室开发了一种新技术，用于创建和研究此类麦克斯韦妖以及用于收集和使用相关信息的相关设备。这种反馈陷阱让我们可以将一个微米级的小珠子放置在一个电位中，我们可以随意设计其形式。这项技术进步在热力学、非平衡统计物理、信息论和控制理论的交叉领域带来了重要的科学应用，特别是 2014 年的一项明确测量证实了一个 60 年前的假设：麦克斯韦妖需要擦除信息，而擦除信息需要消散与获得的能量一样多的能量。未来五年我的研究计划将重点关注三组项目： 1. 异常冷却和加热。  我们最近使用反馈陷阱表明，在相同条件下，热系统有时比热系统冷却得更快（“Mpemba 效应”）。在此成功的基础上，我们将研究最近提出的异常加热场景（逆 Mpemba 效应）和相关问题。 2.信息与热力学实验。  我们将继续致力于信息引擎的研究，总体目标是将麦克斯韦妖从引发悖论的思想实验转变为基本热力学极限的实验说明，并转化为执行有用任务的实际设备。我们将优化以有限速率运行时的基本计算操作（位擦除、逻辑门）的成本，并研究信息发挥作用的新问题，例如稳定不稳定系统所需的最小功率。 3. 单电池内的热流。   我们将开发新的仪器来研究单个生物细胞的产热。长期的动机是创建一种新工具来了解细胞代谢的动态——营养物质如何在单细胞水平上被消耗。短期目标是解决一个显着的科学差异，即预测和测量的细胞内温度变化之间的“105 差距”。前两个项目直接扩展了我的团队之前（非常成功）的工作。第三个是新方向，有可能创建一个新领域（单细胞量热法）并解决我们这个时代突出的科学难题之一。 上述问题是我们理解生物学和生命以及小规模工程的基础。他们将激励并帮助培训加拿大及其他地区的新一代总部人员。

Nearly 150 years ago, James Clerk Maxwell first asked what happens when a "neat-fingered being"-a demon-takes advantage of information to extract useful energy (work) from heat. The Second Law of Thermodynamics should forbid this, but how? The question galled physicists and provoked debate for well over a century. Starting in 2008, my lab developed a new technique for creating and studying such Maxwell demons and related devices for gathering and using relevant information. This feedback trap lets us place a small, micron-scale bead in a potential whose form we can design, essentially at will. This technological advance has led to important scientific applications at the intersection of thermodynamics, nonequilibrium statistical physics, information theory, and control theory, notably a definitive measurement in 2014 that confirms a 60-year-old hypothesis: that Maxwell demons need to erase information and that erasing information requires dissipating as much energy as is gained. The next five years for my research program will focus on three groups of projects: 1. Anomalous cooling and heating.  We recently used our feedback trap to show that a hot system can sometimes cool more quickly than a warm system under identical conditions ("Mpemba effect"). Building on this success, we will investigate a recently proposed anomalous-heating scenario (inverse Mpemba effect) and related questions. 2. Experiments on information and thermodynamics.  We will continue work on information engines, with the overall goal of transforming Maxwell demons from thought experiments raising paradoxes to experimental illustrations of fundamental thermodynamic limits to practical devices that carry out useful tasks. We will optimize the costs of basic computational operations (bit erasure, logic gates) when operated at finite rates and look at new problems where information plays a role, such as the minimum power needed to stabilize an unstable system. 3. Heat flow in single cells.   We will develop new instrumentation for studying heat production from single biological cells. The longer-term motivation is to create a new tool for understanding the dynamics of cellular metabolism-how nutrients are consumed at the single-cell level. The shorter-term goal is to resolve a striking scientific discrepancy, the "105 gap" between predicted and measured intracellular temperature variations. The first two projects directly extend previous (very successful) work in my group. The third is a new direction, with the potential to create a new field (single-cell calorimetry) and resolve one of the outstanding scientific puzzles of our time. The above questions are ones that underlie our understanding of biology and life, as well as engineering at small scales. They will inspire and help train a new generation of HQP, in Canada and beyond.